next previous
Up: An algorithm to

1. Introduction

Research of the post-main-sequence evolution of the globular cluster (GC) stars raises the necessity of stellar photometry in the most inner parts of the clusters. Penetrating into the cluster centers the astronomers aim to improve, statistically, the sample of bright stars occupying the horizontal branch (HB) and the red giant branch (RGB) with its asymptotic features (AGB) and thus to express more reliably some morphological peculiarities of these branches predicted by the existing theories.

Stellar photometry in crowded fields is a fairly sophisticated task and in the last two decades many powerful packages which help astronomers to carry out such kind of photometry have been created - ROMAFOT (Buonanno et al. 1983; Buonanno & Iannicola 1989), DAOPHOT (Stetson 1987) etc. All of them, however, are designed to work under operating systems serving VAX machines or workstations and only few programs exist under DOS for PCs. Having no regular access to such packages we were forced to develope our own computerized procedure for doing stellar photometry in moderately crowded fields using a PC. The basic ideas of such procedures are widely discussed in the literature (Buonanno et al. 1979; Newell 1979; Stetson 1979 etc.). An important step in our reduction procedure is cleaning close stellar companions, a procedure which can be applied in aperture photometry or other kinds of photometry based on modelling stellar images (like a Gaussian fit) of the inner parts of globular clusters. As a target for testing our image processing tools we choose the GC M5 which is an example of a well investigated cluster (Arp 1962; Simoda & Tanikava 1970; Buonanno et al. 1981; Brocato et al. 1995 etc.) and for which we have a good UBV set of short exposure plates. The data processing had been already completed when we got a possibility to use DAOPHOT (version before 1989) - a package especially designed for stellar photometry in crowded fields. A comparison of the results obtained by two different methods, ours and DAOPHOT's, seemed to us very useful as it would allow us to check whether the achieved accuracy depends on the reductions applied or whether the intrinsic accuracy limits of observing technique (photographic emulsion in our case) and the manner of digitization have been reached. The results of this comparison are presented below. Section 2 deals with the observations, digitization, the manner the star images were located, emulsion linearisation and some data filtering procedures we have used. Section 3 describes the algorithm we have developed to clean stellar companions and derive instrumental magnitudes. In Sect. 4 we comment on our derived results, and conclusions about the application of the cleaning procedure are drawn.


next previous
Up: An algorithm to

Copyright by the European Southern Observatory (ESO)
web@ed-phys.fr